Predictive model for mechanised track maintenance over Indian railways: A case study of Jabalpur division-west central railway

Abstract

Track Maintenance has always been a forefront activity in Indian Railways cutting across all Zonal Railways and Divisions and involving multiple agencies and resources. Primarily, there have been two major methods of Track Maintenance over IR: • Manual Maintenance: This Maintenance involves Track Men and Manual Methods with Small Track Machines/Equipment for Manual Maintenance of Track. These Trackmen work in Gangs/Units for Manual Attention to Track of about 80-100m a Day with a Labour Force of 10 Men. This Kind of Maintenance id generally useful for Picking up Slacks and Localized Attentions. • Mechanised Maintenance: This Maintenance involves Track Machines and Associated Machine and Maintenance Staff for Extensive Attention/ Thorough Attention of Track from One End of the Block Section/KM to Other End. Thus, it helps in attending a large block of track in a shorter period with high levels of quality and output. In the latter, the usage of Track Machines and its Deployment is one of the Core Issues associated in providing Quality Input to Track. At Present, the Deployment Plan of the Track Machines in A Division especially Tamping Machines is made on an Annual Basis with heuristics/ historical usage patterns. Sometimes, Inspections and Special Works also attract Track Machines to the various sections of the Divisions. A lot of time and effort is wasted in shifting and prepping machines from one section to other. This is highly unproductive and an inefficient manner of dealing with such a scarce and expensive resource. To add to the menace is that, the frequency of tamping in a section is determined by the historical performance or the ‘as is needed’ basis thereby, overarching the productivity issues of the Track Machines. Therefore, A Robust Analysis is required for arriving at a Two Step Methodology for Determining the Tamping of Track and the Usage/ Deployment of the Track Machines in the Division which is quite independent in deciding its resources. This Paper Aims at • Establishing a Predictive Relationship between the Track Geometry Index (TGI) of Year 0 and Year 1 with Track Structure Variables and Track Geometry Variables combing the Mechanised Input of Track Maintenance. This shall help in Forecasting TGI a Year Before, thereby enabling us to determine Quarterly Tamping Requirements of Various Sections of the Division. • Establishing a Tamping Frequency for Very Poor Track to Very Good Track across the Division and inculcating the same in the Track Machine Deployment Plan. On Analysis, it is found that the TGI is forecasted based on Track Formations, Location of Yard in the Track or Not and Sleeper Density of Track. This is varied across the KMs of Track by means of Mechanised Input captured in the form of Time Difference in Months between the Tamping of Track and Recording of TGI at the End of Year 1. It is also seen that the New Tamping Frequency can be established on the basis of Yards and Formations and the same can be used for preparing the Deployment Plan of the Track Machines after due incorporation of the TGI Required/To Be Maintained at the End of Year 1. Scenario Analysis and Regression Techniques have been incorporated for building the Mechanised Track Maintenance Model leading to a productivity increase of over 30% in 06 Years i.e. About 5% in a Year amounting to Rs. 170 Crores/Year for IR.